Most people today under 30 years old have probably
never seen the mechanics or electronics inside their many personal devices. Everything is so miniaturized
and optimized that if something does go wrong, there is little chance of the owner repairing it. Instead,
the phone, television, stereo, microwave oven, whatever, gets thrown away and a relatively cheap (compared
to paying for a repair) replacement is purchased (or stolen). Besides, if the item was more than two
years old, it was on the verge of obsolescence anyway.
Up until around the early to mid 1980s you had a fair chance of being able to repair an electronic
circuit if trouble arose because at least with commercial products printed circuit boards (PCBs) were
usually 1- or 2-sided and the components still had leads protruding from the sides of the packages.
A $10 Radio Shack soldering iron and some solder wick was sufficient to remove and replace just about
any failed component. Home brew PCBs could be made to nearly the same quality as commercial versions
using a resist ink pen (basically a Magic Marker) and a dish of ferric chloride etchant liquid. A drill
press helped with making holes for the component leads, but a hand drill would get the job done. No
more, though. If you are resourceful enough to get your cellphone or camera open without destroying
it, you will find a very neatly laid out, extremely high density PCB with parts so small you might wonder
how they could work at all. Forget servicing the thing with a soldering iron and a pair of pliers -
you will need at least a hot air wand, a magnifier, tweezers, and, of course, electrostatic discharge
(ESD) preventative gear.
In 1949 when this article appeared in Radio & Television News, printed circuits were
just coming onto the scene. Bakelite, steatite, and ceramic substrates were typically used at the time.
Some processes were already using printed resistors and small-value inductors via silk-screening techniques.
Part 2. A discussion of the techniques and equipment used in making printed circuits for home-built
units (January 1950).
Thanks to Terry W. for providing this article.
Part I. A review of printed circuit techniques. To be concluded next month with on article on how
the experimenter can apply, in a simplified form, printed circuits to home constructed units.
By John T. Frye
A very loud bang announced to the electronic world early in 1945 that printed circuits had moved from
the experimental to the practical stage, for it was at that time that the National Bureau of Standards,
working closely with the Centralab-Division of the Globe Union Company, began mass production on the
tiny radio proximity fuse for mortar shells: a fuse incorporating a complex electronic circuit "printed"
on a thin steatite plate 1 3/4" long by 1 1/4" wide!
This typical group, only a few of the many commercially
built units already produced, is an example of how Centralab's printed circuit audio amplifier has been
received by the industry.
Since that time, the printed circuit has thrust its tentacles into every portion of the electronic
field; and it has miraculously shrunk everything it touched. Hearing aid amplifiers, complete with batteries,
that are smaller than a cigarette package; personal radios that can be cradled in the palm of the hand;
radio and television subassemblies occupying only one-tenth the space needed for conventional assemblies
and requiring one-half as many soldered connections for installation: these are but a few of the achievements
of this new process, and the surface has barely been scratched. Every day sees new applications of this
method by which space is saved, weight is reduced, assembly is simplified, and cost is cut.
Every electronic worker is certain to come in contact with printed circuits in increasing number,
and it is the purpose of this article to prepare him for that contact by making him familiar with the
various methods and techniques by which these circuits are produced commercially and then showing him
how he can develop and experiment with his own printed circuits.
First, it should be clearly understood that the term "printed circuit" covers any reproduction of
an electrical circuit upon an insulating surface by any process. Essentially it changes a bulky three-dimensional
array of electrical parts and conductors into a compact and very nearly two-dimensional arrangement.
An example best shows how this is done:
Suppose we want to build the complete interstage coupling circuit shown in Fig. 2. First, let us redraw
our diagram on a tiny plate of steatite approximately 1" x 3/4". If. Then let us carefully trace out
the heavy lines with a small brush which we have dipped into a "paint" made by mixing fine particles
of silver together with a liquid binder to hold the particles together and a solvent used to make the
mixture thin enough to brush.
Fig. 1. The "Couplate" unit. It contains a complete interstage coupling circuit.
Fig. 2. Diagram of "Couplate." Finished unit measures 1-1/16 x 13/16 x 3/16 in.
Fig. 3. These individual operations show the method used
in preparing a silk screen.
Fig. 4. Silk-screen printing. Paint is forced through the
open mesh of the screen. After the screen is removed. the surface of the base plate is found to be printed
with an exact, sharp-edged, uniformly thick design of the required conductor circuit. A second stencil
can then be used to print the resistors in their proper location.
Fig. 5. Front and rear views of one of the many hearing-aid
amplifiers that are printed on ceramic plates.
Fig. 6. A high temperature oven is used for firing a group
of printed circuits. (Note lack of hand and eye protection)
Fig. 7. Partially completed electronic circuits printed
on steatite plates and cylinders by the silk-screen process. Light lines are silver conductors and inductors;
dark rectangles are resistors; circular disks are ceramic condensers.
Fig. 8. Illustrating the evolution of an audio plate-to-grid
Next, suppose we have several different solutions of finely powdered graphite or lamp-black, a resin
binder, and a solvent. We can experiment with these until we find just the right combination of mixture,
thickness, and length of line needed to produce resistances equal to R1 and R2;
and then we carefully paint in these resistance lines at the proper points between the silver conducting
lines already drawn. Then we place our little plate in an oven and raise the temperature to the point
where our lines of paint will be "fired" directly to the ceramic base, adhering to it with a tensile
strength of 3000 pounds to the square inch. Finally we solder tiny ceramic condensers of the proper
values across the gaps representing the condensers, and then we attach flexible leads to our silver
paint at points 1, 2, 3, and 4. The result is a "printed circuit" that will perform exactly the same
as one using conventional components, but our printed sub-assembly will be no bigger than a postage
stamp and require only four soldered connections to be made by the radio assembly-line operator. A commercial
version of just such a printed circuit is shown in Fig. 1.
Such a manual process, while pointing up the difference between printed and conventional circuits,
obviously could not be adapted to mass production. Various stenciling methods are the answer to producing
more uniform circuits at higher speed, and the silkscreen process is one of the most successful.
In this system, a fine-meshed silk screen is tightly stretched on a wooden frame and covered with
a photosensitive material that becomes insoluble when exposed to strong ultraviolet light. A photographic-positive
mask of the exact shape of the required conducting circuit is placed on top of the screen, which is
then exposed to the rays from an ultraviolet lamp. Finally, the portions of the film protected by the
mask are washed away in cold water, leaving a stencil of the conductor design to be printed. All four
of these steps are clearly illustrated in Fig. 3.
This finished stencil is held securely against the base plate to be printed; and the circuits can
be printed on practically any insulating material, or even on conducting material that has been coated
with a non-conducting film, such as lacquer, and a quantity of silver paint is placed at one end of
the screen. A neoprene bar, or "squeegee," is moved across the top surface, forcing the paint ahead
of it and down through the open mesh of the design, as is shown in Fig. 4. When the screen is removed,
the surface of the plate is found to be printed with an exact, sharp-edged, uniformly-thick design of
the required conductor circuit. A second stencil can be used to print the resistors in their proper
places. The paint is fired to the base exactly as was done before. This process is shown in Fig. 6.
In Fig. 7 are displayed base plates at various stages of completion.
Brushing and stenciling with a silk screen are not the only ways in which the conducting and resistor
paints are applied. For example, a decalcomania, .on which the circuit is printed on a thin flexible
film that can be transferred to the final surface, is useful in applying the circuits to cylindrical
or irregularly-shaped objects. The film is removed by firing.
Most standard printing processes are also used. As a single example, the required design can be raised
on the face of a rubber stamp, and this stamp can be pressed first on a pad of conducting ink and then
on the surface to be printed. Plating of this printed design will increase its conductance if necessary.
In the same way, other printing processes such as engraving, lithographing, and intaglio are also employed.
You old-timers who used to draw your own grid-leaks with a lead pencil were using a form of printed
circuits that still may have possibilities. Pencils having "leads" of varying degrees of conductivity,
or pens filled with conducting inks are being used experimentally. With such devices an experimental
circuit could be drawn and constructed ready for testing all at one and the same operation!
Condensers can be painted, too, by employing silver disks painted on opposite sides of the base plate
so that the plate material becomes the dielectric. If the plate is constructed of high-dielectric material,
condensers of reasonable capacity can be obtained by this method; otherwise, miniature thin-disk ceramic
condensers are often employed by soldering them with a low temperature solder directly to a silvered
area on the base.
Printed inductors are also used, especially in the low-inductance values. Spiral forms are used on
flat bases, although the more conventional forms can be used when the circuit is printed on the tube
envelope or a cylindrical base plate as is shown in Fig. 9. The inductance of a spiral conductor can
be increased by covering it with an insulating layer of lacquer and then painting another spiral right
on top of it and connecting the two in series, painting another spiral on top of that, etc. The distributed
capacity and the Q of the circuit required are the limiting factors to the usefulness of this method.
Placing a layer of magnetic paint, made of a colloidal suspension of powdered magnetic material,
both beneath and above the spiral conductor, with insulating layers serving to protect the turns of
the inductance from shorting. will also increase the inductance.
The spraying of conducting films on insulated surfaces is another method of printing circuits. The
same paints can be used in paint spray guns as for the stenciled-screen process; or molten streams of
metal can be sprayed through locating stencils. Guns are available in which the metal to be sprayed
is fed into the gun in the form of a wire, where it is heated to the melting point by a hydrogen-acetylene
or other flame. Compressed air is used to atomize the molten metal and to drive it on to the work. This
molten material can .be sprayed on wood, Bakelite, plastic, and even ceramic surfaces.
One popular method employs a plastic base plate. This plate is sandblasted through a mask so that
shallow grooves are cut where the conductors are needed. These grooves are sprayed full of molten metal,
after which the surface can be milled, leaving conducting lines that are flush with the surface of the
plastic base plate.
Still another scheme uses an insulated base plate with a thin evaporated coating of conducting metal.
This is covered with a photosensitive film and exposed to light through a mask. The film is developed
so that the portions exposed to light are removed, and the remaining portions, outlining the desired
circuit, resist an abrasive spray so that the protected portions beneath remain intact while the rest
of the metallic coating is cut away by the sand blast.
Another method of producing "printed circuits" is by chemical deposition. This method is not used
much on a commercial basis because of the very thin layers deposited and other technical difficulties,
but it consists essentially of depositing a thin silver coating on a masked surface by the same chemical
methods that are used in silvering mirrors. Increased conductivity can be secured by repeated silvering
or by plating.
Cathode sputtering and evaporation are two other processes for depositing the metallic film. In the
former, the material to be deposited is used as a cathode and the masked base plate is used as the plate
of a temporary vacuum tube. The "plate" is maintained at a high positive potential with respect to the
cathode, and the latter is raised to a volatizing temperature. The metal particles emitted by the cathode
are attracted to and deposited on the base plate through the stencil openings.
The evaporation process is the same except that the plate is not maintained at a high positive potential.
The cathode material is simply heated in the vacuum until it vaporizes on to the work. This permits
the use of non-metallic as well as metallic base plates. In neither case is the film deposited thick
enough to be used for conductors, but this can be overcome by plating.
The radio technician is very familiar with one form of printed circuit: the die-stamped loop antenna.
This is produced by placing a thin sheet of copper on top of a composition or bakelite panel with a
layer of thermoplastic cement between. This sandwich is placed in a punch press, and at one stroke the
metal is cut into a helix and is bonded to the panel.
Dusting is the final major method of printing circuits. This consists of depositing a layer of metallic
dust on a base plate along the lines where conductors or resistors are required and then raising the
temperature sufficiently to drive off the bonding material and to fuse the metal particles together
and to the plate. The entire plate can be covered with an adhesive material and the dust applied through
a stencil, or the adhesive material can be applied through the stencil and then the whole plate subjected
to dusting, with the same results.
Fig. 9. Two complete high-frequency transmitters ready to be connected
to a power supply. The one printed on the glass envelope of the 6J4 tube operates on 136 mc.; that printed
on the ceramic cylinder surrounding the subminiature triode operates on a frequency of 116 mc. Both
transmitters are intended for grid modulation.
While an attempt has been made to touch on all of the methods ordinarily used for printing circuits,
the new industry is advancing so rapidly that one cannot be sure how long this will hold true. Very
recently, for example, the Glass Products Company of Chicago announced a new process, "Micro-screening,"
which they claim has several advantages over the silk-screen methods. Unfortunately, because of current
patent proceedings, details of this new method are not available.
Several illustrations are given to show the wide variety of devices to which printed circuits are
applied. For a more detailed discussion of the various methods discussed in this article, the author
recommends the purchase, for 25c, of "Printed Circuit Techniques," by Cledo Brunetti and Roger W. Curtis.
This National Bureau of Standards Circular 468 can be obtained from the Superintendent of Documents,
U. S. Government Printing Office, Washington. D. C. An excellent group of references for further reading
will be found in the back of this booklet.
Part 2 of this article will be concerned solely with explaining and illustrating how the experimenter
can design and construct his own printed circuits with materials easily obtainable. (To be continued)
Posted May 25, 2013